Reactions on Immobilized Enzymes can Telescope Process Steps, Eliminate Isolations and Reduce Costs at All Stages of Pharmaceutical Development

kilomentor | 17 July, 2012 14:45

At the Current Process Chemistry Conference put on by Cambridge Healthtech Institute in Princeton, June 13-14^th , Mathew Truppo, Process Chemistry Merck & Company, gave a talk about the application of immobilized enzymes to synthesis using the example of the synthesis of the diabetes II drug Januvia (sitagliptin). The use of immobilization allowed the enzymes to be recovered and recycled efficiently and cost effectively. Because the enzymes were easily separated this meant that complicated separation procedures forremoving denatured enzymatic protein from the product were avoided. Because the process was conducted by flowing the reactants past an immobilized plug of enzyme, trapped water phase, and coenzyme there was no need to do a water rich/organic rich phase separation or the disposal of the water-rich phase. Enzyme and coenzyme were immobilized within a solid with structured pores that contained the aqueous fluid.

Sitagliptin is a tertiary amide with a beta amino. It has a single centre of chiralty at the amino bearing carbon. It is produced by stereoselective reductive amination of the beta ketone. Dr. Truppo did not detail the particular immobilization used in this case but from the slides it appeared that the enzyme along with a small portion of its aqueous environment was trapped in some geometrically regular solid pore structure. Cofactors were involved in the transformation and it was stated or implied that these cofactors were also trapped in the immobilizing system. He stated that getting and maintaining the correct water content was very important. Isopropyl amine was continuously supplied in the reactant stream. Acetone the co-product was continuously removed from the system.

This Sitagliptin talk from Dr. Truppo blended with another presentation by Dr. Chris Savile of Codexis pertaining to Codexis’s progress developing a systematic methodology for the evolution of improved catalysts. Both presenters made it clear that now one could afford to use immobilized enzymes from the very earliest stages of development all the way to commercialization, because a client only needed to pay for incremental improvements in the enzyme as one moved a project forward. That is to say, at the very early stages a company like Codexis could provide an enzyme off the shelf, at competitive cost, that could produce small amounts of material; then, when kilograms were needed, the enzyme specialist would perform more optimization to take the enzyme improvement to a stage where it could supply that need; and finally, if the process moved through the clinic into manufacture, another round of refinement could be performed to really optimize the enzyme for full production. Thus one would only need to pay an enzyme specialist collaborator to produce a fit-for-purpose enzyme, and not suffer a fully optimized, full price cost even for candidates that fail as drugs.

Both speakers made the point that one of the areas of cost saving was that with immobilized enzyme processes in a flow system, scaling up was much more predictable. For transformations where many variants of enzyme are available in bulk synthetic chemists should now think first about the possibility to use these enzymes.

Comments

Re: Reactions on Immobilized Enzymes can Telescope Process Steps, Eliminate Isolations and Reduce Costs at All Stages of Pharmaceutical Development

steven alley | 19/07/2012, 08:36

A question on solvates. We are having problems with a crystallisation process. Some batches filter and dry nicely. Others filter nicely but form slurries in the drier (20L rotary evaporator). The batches which slurry have lower yields (by 3-5%) and form lumps which require further processing to remove.

Crystallisation temperature, cooling profile and anti-solvent addition all seem reproducible. Solvent is water. Antisolvent is IPA.

We suspect an IPA solvate is forming in the problem batches. Any suggestions on how we might approach this problem would be gratefully recieved.

«	May 2013					»
Mo	Tu	We	Th	Fr	Sa	Su
		1	2	3	4	5
6	7	8	9	10	11	12
13	14	15	16	17	18	19
20	21	22	23	24	25	26
27	28	29	30	31

kilomentor